acli
Pass
Audited by Gen Agent Trust Hub on Apr 4, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill documents the use of the acli tool for terminal-based Jira operations, including potentially destructive actions such as deleting tickets or attachments. Examples like 'acli jira workitem edit --jql ... --yes' encourage the use of flags that bypass user confirmation, increasing the risk of unintended bulk modifications.\n- [PROMPT_INJECTION]: The skill facilitates indirect prompt injection by retrieving and processing external data from Jira tickets.\n
- Ingestion points: The 'acli jira workitem view' command in SKILL.md loads ticket summaries, descriptions, and comments into the agent's context.\n
- Boundary markers: Absent; ticket content is not demarcated by delimiters or instructions to ignore embedded commands.\n
- Capability inventory: The agent has extensive permissions documented in SKILL.md to modify Jira state, including creating comments, editing tickets, and transitioning statuses.\n
- Sanitization: Absent; there is no instruction for the agent to validate or filter retrieved ticket content for malicious instructions.
Audit Metadata